arXiv subjects

Seungho Lee

Publications and source records attributed to Seungho Lee.

At least 19 recordsLinked to original sources

arXiv2026

Spin-orbit-induced quantum chiral phases

The scalar spin chirality (SSC), whose nonzero value $\langle {\bf S}_i \cdot ({\bf S}_j \times {\bf S}_k) \rangle \neq 0$ implies the breaking of time-reversal and certain point-group symmetries in the ground state, is a key quantity characterizing chiral magnetism in both classical and quantum settings. The classical SSC is manifested, for instance, in skyrmion crystal phase, while the quantum SSC is still highly sought after in various frustrated spin-1/2 models. An interesting possibility that has not been explored so far is the case in which SSC is symmetry-wise allowed, yet remains zero classically due to the collinear or coplanar arrangement of spins, but is generated by virtue of quantum fluctuation. We demonstrate the existence of precisely such a phase in a spin-1/2 triangular-lattice model with XXZ interaction, spin-orbit-induced exchange interactions, and an external magnetic field. Using iDMRG, we thoroughly map out the phase diagram of the model and identify several phases with coexisting magnetic order and SSC. The nonzero SSC arises despite the classical magnetic order being collinear or coplanar. We provide detailed magnon analysis to ascribe its origin to quantum fluctuations around classical magnetic order. The estimate of SSC from magnon analysis agrees with iDMRG results quantitatively. We map out the magnon spectrum and its Berry curvature, culminating in the prediction of finite thermal Hall conductivity in these phases with SSC.

cond-mat.str-el↗

arXiv2026

Projector, Neural, and Tensor-Network Representations of $\mathbb{Z}_N$ Cluster and Dipolar-cluster SPT States

The $\mathbb{Z}_N$ cluster-state wavefunction, a paradigmatic example of symmetry-protected topological (SPT) order with $\mathbb{Z}_N \times \mathbb{Z}_N$ symmetry, is expressed in various equivalent ways. We identify the projector-based scheme called the $P$-representation as the efficient way to express cluster and dipolar cluster state's wavefunctions. Employing the restricted Boltzmann machine scheme to re-write the interaction matrix in the $P$-representation in terms of neural weight matrices allows us to develop the neural quantum state (NQS) and the matrix product state (MPS) representations of the same state. The NQS and MPS representations differ only in the way the weight matrices are split and grouped together in a matrix product. For both $\mathbb{Z}_N$ cluster and dipolar cluster states, we derive in closed form the weight function $W(s,h)$ that couples physical spins $s$ to hidden variables $h$, generalizing the previous construction for $Z_2$ cluster states to $\mathbb{Z}_N$. For the dipolar cluster state protected by two charge and two dipole symmetries, the procedure we have developed leads to the tensor product state (TPS) representation of the wavefunction where each local tensor carries three virtual indices connecting a given site to two nearest neighbors and one further neighbor. We benchmark the resulting TPS construction against conventional MPS representation using density-matrix renormalization group simulations and argue that the TPS could offer a more efficient representation for some modulated SPT states. As a by-product of the investigation, we generalize the previous $Z_2$ matrix product operator construction of the Kramers-Wannier (KW) operator to $\mathbb{Z}_N$ and interprets it as the dipolar generalization of the discrete Fourier transform on $\mathbb{Z}_N$ variables. The new interpretation naturally explains why the KW map is non-invertible.

cond-mat.dis-nn↗

arXiv2025

Dynamics of $\mathrm{CP}^{N-1}$ skyrmions

We derive several exact results for the dynamics of CP$^{N-1}$ skyrmions with arbitrary $N$. Fractonic continuity equation is shown to hold for arbitrary CP$^{N-1}$ fluid implying the conservation of the topological charge and the dipole moment. Inclusion of the Gilbert damping modifies the continuity equation, resulting in the violation of the dipole moment conservation but not of the topological charge. Thiele's equation for the CP$^{N-1}$ skyrmion follows from the modified continuity equation. The Girvin-MacDonald-Platzman (GMP) algebra in the long-wavelength limit is derived for arbitrary CP$^{N-1}$ fluid. In the case of CP$^2$ skyrmions, we identify stable skyrmions in which the ferromagnetic moments are dominant. One can associate a nonzero CP$^1$ charge equal to half the CP$^2$ skyrmion charge and argue that the topological Hall effect of electrons should exist due to their coupling to the ferromagnetic part of the CP$^2$ texture.

cond-mat.str-el↗

arXiv2025

Effective Field Theory of a Noncollinear Altermagnet

We derive an effective field theory for a noncollinear altermagnet and magnons on top of the noncollinear ground state from an altermagnetic Heisenberg model. We obtain the ground-state phase diagram, revealing a noncollinear phase and four distinct collinear phases. The ground state of the noncollinear phase fully breaks the spin rotational symmetry, while the ground state of the collinear phases possesses unbroken $\mathrm{SO}(2)$ symmetry. The resulting effective field theory for the noncollinear phase is an $\mathrm{SO}(3)$ sigma model in which the magnonic excitation has three independent degrees of freedom and exhibits the $d$-wave-like anisotropic linear dispersion. We also discuss possible topological solitons, including $\mathbb{Z}_2$ vortices.

cond-mat.mes-hall↗

arXiv2025

Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentation

Open-Vocabulary Part Segmentation (OVPS) is an emerging field for recognizing fine-grained parts in unseen categories. We identify two primary challenges in OVPS: (1) the difficulty in aligning part-level image-text correspondence, and (2) the lack of structural understanding in segmenting object parts. To address these issues, we propose PartCATSeg, a novel framework that integrates object-aware part-level cost aggregation, compositional loss, and structural guidance from DINO. Our approach employs a disentangled cost aggregation strategy that handles object and part-level costs separately, enhancing the precision of part-level segmentation. We also introduce a compositional loss to better capture part-object relationships, compensating for the limited part annotations. Additionally, structural guidance from DINO features improves boundary delineation and inter-part understanding. Extensive experiments on Pascal-Part-116, ADE20K-Part-234, and PartImageNet datasets demonstrate that our method significantly outperforms state-of-the-art approaches, setting a new baseline for robust generalization to unseen part categories.

cs.CV↗

arXiv2025

Learning to Transfer Human Hand Skills for Robot Manipulations

We present a method for teaching dexterous manipulation tasks to robots from human hand motion demonstrations. Unlike existing approaches that solely rely on kinematics information without taking into account the plausibility of robot and object interaction, our method directly infers plausible robot manipulation actions from human motion demonstrations. To address the embodiment gap between the human hand and the robot system, our approach learns a joint motion manifold that maps human hand movements, robot hand actions, and object movements in 3D, enabling us to infer one motion component from others. Our key idea is the generation of pseudo-supervision triplets, which pair human, object, and robot motion trajectories synthetically. Through real-world experiments with robot hand manipulation, we demonstrate that our data-driven retargeting method significantly outperforms conventional retargeting techniques, effectively bridging the embodiment gap between human and robotic hands. Website at https://rureadyo.github.io/MocapRobot/.

cs.RO↗

arXiv2024

MaskRIS: Semantic Distortion-aware Data Augmentation for Referring Image Segmentation

Referring Image Segmentation (RIS) is an advanced vision-language task that involves identifying and segmenting objects within an image as described by free-form text descriptions. While previous studies focused on aligning visual and language features, exploring training techniques, such as data augmentation, remains underexplored. In this work, we explore effective data augmentation for RIS and propose a novel training framework called Masked Referring Image Segmentation (MaskRIS). We observe that the conventional image augmentations fall short of RIS, leading to performance degradation, while simple random masking significantly enhances the performance of RIS. MaskRIS uses both image and text masking, followed by Distortion-aware Contextual Learning (DCL) to fully exploit the benefits of the masking strategy. This approach can improve the model's robustness to occlusions, incomplete information, and various linguistic complexities, resulting in a significant performance improvement. Experiments demonstrate that MaskRIS can easily be applied to various RIS models, outperforming existing methods in both fully supervised and weakly supervised settings. Finally, MaskRIS achieves new state-of-the-art performance on RefCOCO, RefCOCO+, and RefCOCOg datasets. Code is available at https://github.com/naver-ai/maskris.

cs.CV↗

arXiv2024

Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation

We address the challenges of the semi-supervised LiDAR segmentation (SSLS) problem, particularly in low-budget scenarios. The two main issues in low-budget SSLS are the poor-quality pseudo-labels for unlabeled data, and the performance drops due to the significant imbalance between ground-truth and pseudo-labels. This imbalance leads to a vicious training cycle. To overcome these challenges, we leverage the spatio-temporal prior by recognizing the substantial overlap between temporally adjacent LiDAR scans. We propose a proximity-based label estimation, which generates highly accurate pseudo-labels for unlabeled data by utilizing semantic consistency with adjacent labeled data. Additionally, we enhance this method by progressively expanding the pseudo-labels from the nearest unlabeled scans, which helps significantly reduce errors linked to dynamic classes. Additionally, we employ a dual-branch structure to mitigate performance degradation caused by data imbalance. Experimental results demonstrate remarkable performance in low-budget settings (i.e., <= 5%) and meaningful improvements in normal budget settings (i.e., 5 - 50%). Finally, our method has achieved new state-of-the-art results on SemanticKITTI and nuScenes in semi-supervised LiDAR segmentation. With only 5% labeled data, it offers competitive results against fully-supervised counterparts. Moreover, it surpasses the performance of the previous state-of-the-art at 100% labeled data (75.2%) using only 20% of labeled data (76.0%) on nuScenes. The code is available on https://github.com/halbielee/PLE.

cs.CV↗

arXiv2024

Stable singular fractional skyrmion spin texture from the quantum Kelvin-Helmholtz instability

Topology profoundly influences diverse fields of science, providing a powerful framework for classifying phases of matter and predicting nontrivial excitations, such as solitons, vortices, and skyrmions. These topological defects are typically characterized by integer numbers, called topological charges, representing the winding number in their order parameter field. The classification and prediction of topological defects, however, become challenging when singularities are included within the integration domain for calculating the topological charge. While such exotic nonlinear excitations have been proposed in the superfluid $^3$He-A phase and spinor Bose-Einstein condensate of atomic gases, experimental observation of these structures and studies of their stability have long been elusive. Here we report the observation of a singular skyrmion that goes beyond the framework of topology in a ferromagnetic superfluid. The exotic skyrmions are sustained by undergoing anomalous symmetry breaking associated with the eccentric spin singularity and carry half of the elementary charge, distinctive from conventional skyrmions or merons. By successfully realizing the universal regime of the quantum Kelvin-Helmholtz instability, we identified the eccentric fractional skyrmions, produced by emission from a magnetic domain wall and a spontaneous splitting of an integer skyrmion with spin singularities. The singular skyrmions are stable and can be observed after 2~s of hold time. Our results confirm the universality between classical and quantum Kelvin-Helmholtz instabilities and broaden our understanding on complex nonlinear dynamics of nontrivial texture beyond skyrmion in topological quantum systems.

cond-mat.quant-gas↗

arXiv2024

Domain Wall Networks as Skyrmion Crystals in Chiral Magnets

We theoretically investigate the ground states of a chiral magnet with a square anisotropy and show that it supports domain wall networks as stable ground states. A domain wall junction in the domain wall network turns out to be a skyrmion with half topological charge and, therefore, the found domain wall network has a second topological nature, a skyrmion crystal. More specifically, we present a ground-state phase diagram of the chiral magnet with varying anisotropy parameters consisting of skyrmion lattices, chiral soliton lattices, and ferromagnetic states. In the presence of the square anisotropy, the skyrmion crystal forms a domain wall network. The size of domains in the domain wall network is shown to be tunable by an external magnetic field, offering a way to realize experimentally detectable domain wall networks.

cond-mat.mes-hall↗

arXiv2024

Understanding Multi-Granularity for Open-Vocabulary Part Segmentation

Open-vocabulary part segmentation (OVPS) is an emerging research area focused on segmenting fine-grained entities using diverse and previously unseen vocabularies. Our study highlights the inherent complexities of part segmentation due to intricate boundaries and diverse granularity, reflecting the knowledge-based nature of part identification. To address these challenges, we propose PartCLIPSeg, a novel framework utilizing generalized parts and object-level contexts to mitigate the lack of generalization in fine-grained parts. PartCLIPSeg integrates competitive part relationships and attention control, alleviating ambiguous boundaries and underrepresented parts. Experimental results demonstrate that PartCLIPSeg outperforms existing state-of-the-art OVPS methods, offering refined segmentation and an advanced understanding of part relationships within images. Through extensive experiments, our model demonstrated a significant improvement over the state-of-the-art models on the Pascal-Part-116, ADE20K-Part-234, and PartImageNet datasets.

cs.CV↗

arXiv2024

Observation of Analogue Dynamic Schwinger Effect and Non-Perturbative Light Sensing in Lead Halide Perovskites

Dielectric breakdown of physical vacuum (Schwinger effect) is the textbook demonstration of compatibility of Relativity and Quantum theory. Although observing this effect is still practically unachievable, its analogue generalizations have been shown to be more readily attainable. This paper demonstrates that a gapped Dirac semiconductor, methylammonium lead-bromide perovskite (MAPbBr$_3$), exhibits analogue dynamical Schwinger effect. Tunneling ionization under deep sub-gap mid-infrared irradiation leads to intense photoluminescence in the visible range, in full agreement with quasi-adiabatic theory. In addition to revealing a gapped extended system suitable for studying the analogue Schwinger effect, this observation holds great potential for non-perturbative field sensing, i.e., sensing electric fields through non-perturbative light-matter interactions. First, this paper illustrates this by measuring the local deviation from the nominally cubic phase of a perovskite single crystal, which can be interpreted in terms of frozen-in fields. Next, it is shown that analogue dynamic Schwinger effect can be used for nonperturbative amplification of non-parametric upconversion process in perovskites driven simultaneously by multiple optical fields. This discovery demonstrates the potential for material response beyond perturbation theory in the Schwinger regime, offering extremely sensitive light detection and amplification across an ultrabroad spectral range not accessible by conventional devices.

cond-mat.mes-hall↗

arXiv2024

Self-Supervised Vision Transformers Are Efficient Segmentation Learners for Imperfect Labels

This study demonstrates a cost-effective approach to semantic segmentation using self-supervised vision transformers (SSVT). By freezing the SSVT backbone and training a lightweight segmentation head, our approach effectively utilizes imperfect labels, thereby improving robustness to label imperfections. Empirical experiments show significant performance improvements over existing methods for various annotation types, including scribble, point-level, and image-level labels. The research highlights the effectiveness of self-supervised vision transformers in dealing with imperfect labels, providing a practical and efficient solution for semantic segmentation while reducing annotation costs. Through extensive experiments, we confirm that our method outperforms baseline models for all types of imperfect labels. Especially under the zero-shot vision-language-model-based label, our model exhibits 11.5\%p performance gain compared to the baseline.

cs.CV↗

arXiv2023

Weakly Supervised Semantic Segmentation for Driving Scenes

State-of-the-art techniques in weakly-supervised semantic segmentation (WSSS) using image-level labels exhibit severe performance degradation on driving scene datasets such as Cityscapes. To address this challenge, we develop a new WSSS framework tailored to driving scene datasets. Based on extensive analysis of dataset characteristics, we employ Contrastive Language-Image Pre-training (CLIP) as our baseline to obtain pseudo-masks. However, CLIP introduces two key challenges: (1) pseudo-masks from CLIP lack in representing small object classes, and (2) these masks contain notable noise. We propose solutions for each issue as follows. (1) We devise Global-Local View Training that seamlessly incorporates small-scale patches during model training, thereby enhancing the model's capability to handle small-sized yet critical objects in driving scenes (e.g., traffic light). (2) We introduce Consistency-Aware Region Balancing (CARB), a novel technique that discerns reliable and noisy regions through evaluating the consistency between CLIP masks and segmentation predictions. It prioritizes reliable pixels over noisy pixels via adaptive loss weighting. Notably, the proposed method achieves 51.8\% mIoU on the Cityscapes test dataset, showcasing its potential as a strong WSSS baseline on driving scene datasets. Experimental results on CamVid and WildDash2 demonstrate the effectiveness of our method across diverse datasets, even with small-scale datasets or visually challenging conditions. The code is available at https://github.com/k0u-id/CARB.

cs.CV↗

arXiv2022

Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Report

Image super-resolution is a common task on mobile and IoT devices, where one often needs to upscale and enhance low-resolution images and video frames. While numerous solutions have been proposed for this problem in the past, they are usually not compatible with low-power mobile NPUs having many computational and memory constraints. In this Mobile AI challenge, we address this problem and propose the participants to design an efficient quantized image super-resolution solution that can demonstrate a real-time performance on mobile NPUs. The participants were provided with the DIV2K dataset and trained INT8 models to do a high-quality 3X image upscaling. The runtime of all models was evaluated on the Synaptics VS680 Smart Home board with a dedicated edge NPU capable of accelerating quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 60 FPS rate when reconstructing Full HD resolution images. A detailed description of all models developed in the challenge is provided in this paper.

eess.IV↗

arXiv2022

Magnon dynamics in a Skyrmion-textured domain wall of antiferromagnets

We theoretically investigate the interaction between magnons and a Skyrmion-textured domain wall in a two-dimensional antiferromagnet and elucidate the resultant properties of magnon transport. Using supersymmetric quantum mechanics, we solve the scattering problem of magnons on top of the domain wall and obtain the exact solutions of propagating and bound magnon modes. Then, we find their properties of reflection and refraction in the Skyrmion-textured domain wall, where magnons experience an emergent magnetic field due to its non-trivial spin texture-induced effective gauge field. Based on the obtained scattering properties of magnons and the domain wall, we show that the thermal transport decreases as the domain wall's chirality increases. Our results suggest that the thermal transport of an antiferromagnet is tunable by modulating the Skyrmion charge density of the domain wall, which might be useful for realizing electrically tunable spin caloritronic devices.

cond-mat.mes-hall↗

arXiv2021

Laser astrophysics experiment on the amplification of magnetic fields by shock-induced interfacial instabilities

Laser experiments are becoming established as a new tool for astronomical research that complements observations and theoretical modeling. Localized strong magnetic fields have been observed at a shock front of supernova explosions. Experimental confirmation and identification of the physical mechanism for this observation are of great importance in understanding the evolution of the interstellar medium. However, it has been challenging to treat the interaction between hydrodynamic instabilities and an ambient magnetic field in the laboratory. Here, we developed an experimental platform to examine magnetized Richtmyer-Meshkov instability (RMI). The measured growth velocity was consistent with the linear theory, and the magnetic-field amplification was correlated with RMI growth. Our experiment validated the turbulent amplification of magnetic fields associated with the shock-induced interfacial instability in astrophysical conditions for the first time. Experimental elucidation of fundamental processes in magnetized plasmas is generally essential in various situations such as fusion plasmas and planetary sciences.

physics.plasm-ph↗

arXiv2021

Railroad is not a Train: Saliency as Pseudo-pixel Supervision for Weakly Supervised Semantic Segmentation

Existing studies in weakly-supervised semantic segmentation (WSSS) using image-level weak supervision have several limitations: sparse object coverage, inaccurate object boundaries, and co-occurring pixels from non-target objects. To overcome these challenges, we propose a novel framework, namely Explicit Pseudo-pixel Supervision (EPS), which learns from pixel-level feedback by combining two weak supervisions; the image-level label provides the object identity via the localization map and the saliency map from the off-the-shelf saliency detection model offers rich boundaries. We devise a joint training strategy to fully utilize the complementary relationship between both information. Our method can obtain accurate object boundaries and discard co-occurring pixels, thereby significantly improving the quality of pseudo-masks. Experimental results show that the proposed method remarkably outperforms existing methods by resolving key challenges of WSSS and achieves the new state-of-the-art performance on both PASCAL VOC 2012 and MS COCO 2014 datasets.

cs.CV↗